Support NeoGAF

[In Cold Blood]

We all know that there are proprietary engines, and then third party engines like Unreal and Unity.
We also all know that not all engines are created equal, and not all engines are suitable for every type of game.
Some are awesome for FPS, like Frostebite, but then when it was mandated to be used by all studios, it didn't perform as well.
There will always be a need for some studios to maintain their own engines.
Some of the best looking games are actually on proprietary engines. God of War, The Last Of Us, Ratchet and Clank, Horizon, Doom, Forza, Plague Tale and Red Dead Redemption for instance, which begs the question, why would they change?

There are reasons why some studios will ditch their own game engine and switch to Unreal for instance. It's expensive to update an engine. Not just any developer can build an engine, let alone a good one.
So for some, like CDPR, it's just not worth the trouble. The time it takes to update an engine can take 12-18 months, and that's down time for alot of the studio. Maybe they don't have the skillset to keep up?
Now some smaller studios have shown that they are able to compete. Asobos engine for instance can produce games as good looking as any, and they are a smaller studio in the scheme of things.

There is, however, another major issue that studios face that might influence their decision to go with Unreal over their own engine, and that is attracting talent. Top talent.

Every single developer has worked on Unreal Engine at some point. It's a good engine to use, and becoming proficient on it can help you in getting a good job.
So in a world where the best devs are in.high demand, they can be picky about what jobs they want to take.
A very important thing for a developer is his resume, or body of work. It's one thing to say you worked on an idie game, it's another to say you worked on The Last Of Us.

So herein lies the issues for studios, and one that a few have recently had to deal with.
Take 343. They brought out a new engine for Halo. It was by all accounts, a bit of a dog. It took devs awhile to get their heads around, and as MS had a maximum time limit of 18 months for contractors, they would then have to move them on and bring on another newbie that then had to get used to the engine.
For 343, using Slipspace became a issue for them in attracting devs to come and work for them. If you are a dev and you spend 18 months working on Slipspace, and then have to move on and none of that experience you got with that engine is going to help you in any other game development studio, why would you bother?

Now, some studios will always be attractive to quality devs regardless of them using a different engine.or not. Sure, learning how to work on ND, or SSM, or Id or Rockstars engine won't help you in any other studio, having on your resume that you worked on GOW, or Red Dead Redemption is very attractive for potential employers.

This is one of the major issues facing studios using their own engines. Devs love Unreal. They would rather work on Unreal then some smaller engine you won't ever use again.
As Unreal is the most used engine in the world, your bread is best buttered by being a legend on that.

This is becoming one of the major issues facing studios when deciding if to keep using their own engine, or going with Unreal.

 

[HL3.exe]

As someone who's been pretty well versed in game-engines and middleware technologies over the years, I've come to the conclusion that I personally give less of a fuck, as long as the game itself is good and that the devs have the proper tools to make it work for them.

These days complex game-logic is less hampered by game-engines, but more about the implementation of ideas and how flexible the tools are to work with for devs (to prototype these concepts/ideas). We've hit diminishing returns on visuals for a while now. The only real reason devs keep there own tools is to lower cost or have specific needs for their game-logic.

 

[Drizzlehell]

The OP is emphasizing the challenge studios face in attracting talented developers when deciding whether to continue using their own game engine or adopt Unreal Engine. They highlight the preference of developers for Unreal Engine due to its industry recognition and the value it adds to their resumes, making it harder for studios to hire skilled developers if they stick with their proprietary engines. Ultimately, the OP suggests that studios should consider the talent attraction factor when making decisions about game engines.

 

[Fess]

Considering how problematic it seems to be to get stable performance in Unreal Engine even on a powerful PC I have nothing but concern for the future of studios moving over to UE. And it seems like many studios have trouble getting games out the door on UE5. How many are out now?

 

[cireza]

Metroid Prime Remastered runs like it does on Switch because it is not using a stupid third party engine.

Dead or Alive 5+ runs like it does on Vita because it is not using a stupid third party engine.

SMT V runs like it does on Switch because it is using Unreal Engine.

 

[Fafalada]

Devs love Unreal.

This is actually a relatively rare sentiment - to clarify, anyone that's spent any substantial amount actually shipping software with 'an engine' - is rarely going to profess love for it - regardless of which engine we're talking about (includes people that used their own).
It's much easier to be a fan of one if you haven't gone through a real shipping exercise with it though - and Unreal's toybox is quite appealing there.

As Unreal is the most used engine in the world

It's not even close - though they've made some gains in last 2 years, the gap to the market leader is still massive (4:1). But you're entirely right that from perspective of 'how do I get people up to speed / hire them out of the box with experience' 3rd party engines have a distinct advantage (Unreal still has millions of users, so nothing proprietary will ever come close). But that's been the case for roughly 20 years or so now, it's not a new or even recent phenomenon.

Though the flipside of this is the same as it has been for eg. software engineers for past 40 something years - best SEs are not hired on 'what's your C++ expertise' - they are problem solvers that can quickly leverage and adopt/learn many different tools&technologies, picking the best for the job. Knowledge of particular technology can be really useful when you're doing fast-hiring ramp-up, but it's almost a non-decision factor when you're just searching for the best talent.

 

[TrueLegend]

Ubisoft New Engine Update to Snowdrop Rocks: That Avatar and especially that Star Wars Game look lit

RE engine looks lit and with Pragmata we may get proper implementation of RT features in RE engine

Rockstar is not going to switch to unreal as RDR 2 still looks nuts

Call of Duty is looking and running great.

Microsoft has ID

Sony studios like Naughty Dogs, Santa Monica, Guerrilla games are doing great.
R&C, Demon's Souls and HFW are looking great

All this to put in perspective that the idea of UE5 of being one engine to rule them all is greatly exaggerated. Also, the engine has many CPU side bottlenecks.

 

[winjer]

The issue is that when it comes to third party engines, we are stuck with UE5.
Some companies have their in-house engines, like Frostbite, RE Engine, Snowdrop, etc.
But for most studios, they have to use UE5. Unity is used by some studios, but it's mostly for indie games and mobiles.
There will be a lot of problems if we have only one major game engine in the industry.
Lack of competition, eventually leads to higher prices and less innovation.
If UE5 has some issue, then a ton of games will also have that issue. This has been a sore problem with UE4, which has issues with stutters from shader compilation and asset streaming. And because of that, a ton of games are being released with these issues.
UE4 also has problems with using multiple threads, being mostly limited by one main thread. And UE5 seems to have that same problem.

 

[Wildebeest]

Unreal Engine gets a lot of hype and a lot of student project looking demos, but I don't think it is quite the reliable workhorse it is made out to be. At least not without a lot of work by developers who really know how to get the best from it.

 

[Msamy]

expect the matrix demo, i don't see any released unreal engine 5 game that i can say its the best looking game, and currently the 2 best looking games for me are burning shores which use Decima engine and cyberpunk 2077 overdrive which use red engine .

 

[Fess]

Unreal Engine gets a lot of hype and a lot of student project looking demos, but I don't think it is quite the reliable workhorse it is made out to be. At least not without a lot of work by developers who really know how to get the best from it.

Yeah. The Coalition are considered fairly skilled Unreal Engine devs, and they did the deeper test they showed at GDC if they could reach stable 60fps at high fidelity, they didn’t quite get there but still thought it would be possible with engine updates and optimization. Then several MS studios swapped over to UE5.

Fast forward 2 years. Zero updates on what The Coalition is working on, not even a teaser at this year’s Xbox showcase. Hellblade 2 is still shown with cutscenes. Nothing from State of Decay 3. Perfect Dark, nothing. Not sure what’s going on at 3rd party studios but there is nothing new from Stalker 2 and Black Myth is missing.

What UE5 games has shipped so far?

 

[EverydayBeast]

After tears of the kingdom Nintendo proved they’re top of the food chain engine wise.

 

[brian0057]

I want a AAA videogame coded and compiled entirely in Assembly.

 

[RoadHazard]

This is actually a relatively rare sentiment - to clarify, anyone that's spent any substantial amount actually shipping software with 'an engine' - is rarely going to profess love for it - regardless of which engine we're talking about (includes people that used their own).
It's much easier to be a fan of one if you haven't gone through a real shipping exercise with it though - and Unreal's toybox is quite appealing there.


It's not even close - though they've made some gains in last 2 years, the gap to the market leader is still massive (4:1). But you're entirely right that from perspective of 'how do I get people up to speed / hire them out of the box with experience' 3rd party engines have a distinct advantage (Unreal still has millions of users, so nothing proprietary will ever come close). But that's been the case for roughly 20 years or so now, it's not a new or even recent phenomenon.

Though the flipside of this is the same as it has been for eg. software engineers for past 40 something years - best SEs are not hired on 'what's your C++ expertise' - they are problem solvers that can quickly leverage and adopt/learn many different tools&technologies, picking the best for the job. Knowledge of particular technology can be really useful when you're doing fast-hiring ramp-up, but it's almost a non-decision factor when you're just searching for the best talent.

What's the most used one? Unity? Mostly with indies I think? And of course indie games outnumber big AAA projects by several orders of magnitude. But among all big studios, which I think is what OP is mostly talking about here, I'm pretty sure Unreal is the undisputed #1?

 

[Kumomeme]

from what i understand, using widely used and popular engine like Unreal Engine also allow studios/company to easier expand their reach toward skilled devs as plenty of people out there utilizing similliar tools as them. it also would smooth out their development because devs can adapt faster due to familliar tools.

 

[Fess]

I want a AAA videogame coded and compiled entirely in Assembly.

Have you seen Jonathan Blow’s talk about preventing the civilization’s collapse?

I’m not sure there are people around that can do low level coding much longer. Once the gray guys that grew up coding C64 games and demos are gone we’ll be stuck iterating on old engines nobody young have a clue how to properly optimize.

 

[Wildebeest]

What UE5 games has shipped so far?

Is that a rhetorical question? Fortnite and Redfall are probably the most high profile.

 

[Three]

Bespoke engines for a particular type of game would always be more efficient because they're built for that type of game with very specialised tricks to optimise and without any baggage. UE is built for ease and familiarity. You just have to choose which you want.

 

[ScHlAuChi]

Having worked with every version of Unreal Engine and being in the games industry for over 20 years I can tell you exactly why studios like CDPR switch to Unreal.
It is tools and productivity. Productivity is the most important factor to drive down production costs of a modern AAA game.
Engine technology doesnt matter much anymore unless you do a game that requires something very specific - like Teardown for example.
Epic has hundreds of devs working on the engine and tools alone. No pure game dev studio can afford to do the same.
It still requires alot of effort and experience to do a GOOD game with UE - an enigne is only as good as the dev using it!

 

[ChorizoPicozo]

you write a lot to end up with a very narrow scope of the actual circumstances around this topic.

you didn't even talk about the licences feeds of a Third Party Engine.

if a proprietary engine has good documentation and support a studio/publisher can ease the integration of new talent..

attracting talent is not an Engine issue. is a culture/location issue.

your halo example proves to previous point. 343 didn't have an issue attracting talent; it was their bullshit work policies that forced acquired talent away.

"Devs love Unreal" I don't think all devs love it. especially Graphics/Software engineers that want to make their own shit.

developing your own technology is not just a status/proud/big dick energy thing...if you want to squeeze all the performance possible of an specific hardware, if you want to maximize profits, have the chance the expedite game dev (not waiting for certain aspects of a brand new engine to be online) or your lack the workforce/talent of specific people to tackle the development of a proprietary engine....this is where a third party engine can become the only option.

 

[Imru’ al-Qays]

The 18 month time limit for contractors thing at Microsoft is just so wild to me. How was nobody at Xbox able to look at that and say "this is fucking stupid, we need to secure an exemption"?

 

[Moochi]

Fortnite makes me wary of UE5. It runs like crap even when "locked" at 144 fps. Stuttering everywhere. Not a pleasant experience.

 

[Topher]

I want a AAA videogame coded and compiled entirely in Assembly.

Think this is as close as you'll get....

.....until AIs start making video games.

 

[Heimdall_Xtreme]

Konami should release Fox - Engine for everyone to use.

That Engine and RE-Engine are the best I've ever seen.

 

[brian0057]

Think this is as close as you'll get....

.....until AIs start making video games.

I think Roller Coaster Tycoon was also made entirely in Assembly.
I know the creator is a programming genius.

 

[Topher]

I think Roller Coaster Tycoon was also made entirely in Assembly.
I know the creator is programming genius.

Wow.....that's impressive.

 

[genesisblock]

Think this is as close as you'll get....

.....until AIs start making video games.

If we're gonna use the term AAA, then we should compare it relative to the generation of consoles. In that case, NES, SNES, Sega Genesis "mega-hits" are "AAA" games programmed in Assembly.

Although that Roller Coaster Tycoon is an insane example.

 

[Topher]

If we're gonna use the term AAA, then we should compare it relative to the generation of consoles. In that case, NES, SNES, Sega Genesis "mega-hits" are "AAA" games programmed in Assembly.

Although that Roller Coaster Tycoon is an insane example.

Pong was used in jest, but you make a good point. Hard to wrap my head around someone developing a game like RCT in Assembly.

 

[Neilg]

The OP reads like they asked chatgpt to write a post about engines. It says a lot but also absolutley nothing

 

[peronmls]

Unreal Engine = All engine and no talent. Just a showcase engine. It's so easy to do stuff in it.